Description: ryan Pic

Bryan Marker

Postdoctoral researcher in the Department of Computer Science at The University of Texas at Austin


E-mail: bamarker@cs.utexas.edu
Desk: GDC 5.702B

Mail: 1 University Station C0500, Austin, TX 78712-1188

Ph.D. in Computer Science

The University of Texas at Austin


B.S. in Computer Science, Turing Scholars honors
B.S. in Mathematics, Scientific Computing option
The University of Texas at Austin



Areas of Interest

high-performance computing, parallel computing, scientific computing, dense linear algebra, software engineering, automatic program generation

 

 

Curriculum Vitae

 

 

My Research

 

My research focuses on developing a new approach to software development called Design by Transformation (DxT). In short, DxT codifies software design knowledge for a domain as transformations with the purpose of automatic program generation.  DxTer is a prototype to generate code automatically via the DxT approach.  For now, I am focusing on the domains of dense linear algebra and tensor contractions.

 

Here is a preprint of a position for SE-CSE.  My advisors and I talk about some of the main points of my research now and its potential in the future.

 

Here are the slides from a keynote Don Batory and I presented at SLE / GPCE 2013.  We describe how “dark knowledge” can be encoded in graph grammars to make explicit the knowledge used to develop code.

 

Here is my dissertation.

 

 

Selected Papers

 

Bryan Marker, Don Batory, and Robert van de Geijn. “Understanding Performance Stairs: Elucidating Heuristics.” International Conference on Automated Software Engineering 2014. Draft: http://www.cs.utexas.edu/ftp/predator/14ASE.pdf

 

Bryan Marker, Tyler Smith, Don Batory, Field Van Zee, and Robert van de Geijn. “Code Generation to Aid Parallel Code Generation.” UT CS TR-08.

 

Bryan Marker, Don Batory, and Robert van de Geijn. “A Case Study in Mechanically Deriving Dense Linear Algebra Code.” The International Journal of High Performance Computing Applications.  Volume 27, Issue 4, November 2013.

 

Bryan Marker, Don Batory, and Robert van de Geijn. “Code Generation and Optimization of Distributed-Memory Dense Linear Algebra Kernels.” iWAPT 2013.  Draft: http://www.cs.utexas.edu/users/flame/pubs/iWAPT13.pdf

 

Bryan Marker, Jack Poulson, Don Batory, and Robert van de Geijn. “Designed Linear Algebra Algorithms by Transformation: Mechanizing the Expert Developer.” iWAPT 2012. Draft: http://www.cs.utexas.edu/users/flame/pubs/iWAPT12.pdf